system

Abstract

We provide the system. [Solution] A sound processing means that monitors the user's voice environment and collects sound data, A data conversion means for converting collected acoustic data into textual information, An analytical method for comparing converted text information with existing fraud patterns to determine the likelihood of fraud, A notification system that sends a warning to pre-registered contacts when a potential scam is detected, A system that includes this.

Description

Technical Field

[0001] The technology of the present disclosure relates to a system.

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, the method including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance as a response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] The damage caused by fictitious charge claim fraud targeting mainly the elderly is increasing socially, and in particular, methods of illegally obtaining personal information and funds through phone calls or direct conversations are rampant. Against such fraudulent acts, there is a need for an effective preventive measure to detect them with high accuracy and in real time and prevent damage.

Means for Solving the Problems

[0005] This invention detects potential fraud by providing an acoustic processing device that monitors user conversations and collects audio data. It identifies fraudulent activity by using a data conversion unit to convert the audio data into text, and further providing a comparison device that compares the converted text with existing fraud patterns. When potential fraud is detected, it provides a means to prevent the spread of damage by promptly sending a notification to family members or public institutions via an alarm unit.

[0006] A "user" is an individual who uses the system and is the person to whom the voice data is collected.

[0007] An "acoustic processing device" is a device that has the function of acquiring the user's voice in real time and capturing it as digital audio data.

[0008] The "data conversion unit" is a component that has the function of converting acquired audio data into text format.

[0009] A "comparison device" is a device used to analyze the likelihood of fraudulent activity by comparing text data with existing fraud patterns.

[0010] The "alarm transmission unit" is a component that has the function of sending a notification to the user's family or a public institution when a risk of fraud is detected.

[0011] A "system" is a collection of devices or programs that include an acoustic processing unit, a data conversion unit, a comparison unit, and an alarm signaling unit, and that work together in coordination. [Brief explanation of the drawing]

[0012] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3]It is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] It is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] It shows an emotion map to which multiple emotions are mapped. [Figure 10] It shows an emotion map to which multiple emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when an emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when an emotion engine is combined.

Embodiments for Carrying Out the Invention

[0013] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0014] First, the terms used in the following description will be explained.

[0015] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0016] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0017] In the following embodiments, the numbered storage is one or more non-volatile storage devices that store various programs, various parameters, and the like. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, and the like.

[0018] In the following embodiments, the numbered communication I / F (Interface) is an interface including a communication processor, an antenna, and the like. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0019] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0020] [First Embodiment]

[0021] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0022] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0023] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0024] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0025] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0026] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0027] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0028] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0029] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0030] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0031] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0032] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0033] This invention relates to an audio data monitoring system for protecting users, including the elderly, from fraud. The system consists of an acoustic processing unit, a data conversion unit, a comparison unit, and an alarm issuing unit.

[0034] First, the device constantly monitors the sounds around the user and uses an acoustic processing unit to collect conversations. Once the audio data is collected, a data conversion unit within the system receives it and converts it from audio to text data. During this conversion process, speech recognition technology generates highly accurate text.

[0035] Next, the server retrieves the converted text data and compares it to a database of previously collected fraudulent activity. This process utilizes AI algorithms to analyze and identify potential fraud in real time. In particular, it is designed to effectively detect specific phrases and contexts that may be associated with fraud.

[0036] If a fraudulent activity is deemed highly likely, the server controls the alarm system to send an alert to the user's family and pre-registered public institutions. This alert includes details about the potential fraud and the user's current location, allowing for a swift response.

[0037] As a concrete example of implementation, consider a scenario where a user receives a suspicious phone call and it is identified as a fraudulent billing attempt. In this case, the device monitors the conversation, processes the necessary information, and immediately warns the family, allowing the user to take steps to prevent becoming a victim of fraud.

[0038] In this way, this system can dynamically and quickly detect fraud using voice information and send alerts to relevant parties, thereby preventing fraud from occurring.

[0039] The following describes the processing flow.

[0040] Step 1:

[0041] The device monitors the sounds around the user in real time and collects them as audio data. Using an acoustic processing device, it reduces ambient noise while clearly capturing the content of conversations.

[0042] Step 2:

[0043] The terminal transmits the collected audio data to the data conversion unit. Here, the audio data is converted into text data using speech recognition software. The converted text is processed to accurately represent the conversation content without being affected by pronunciation quirks or noise.

[0044] Step 3:

[0045] The server receives the text data and applies an AI algorithm to compare it with a fraud database. The AI ​​uses machine learning models to identify unknown tactics as well as matching against known fraud patterns. At this stage, the likelihood of fraud is analyzed.

[0046] Step 4:

[0047] If the server determines that there is a high risk of fraud, it sends information to the alerting unit. The notification is then generated to include details of the suspected fraud and the user's location information.

[0048] Step 5:

[0049] The server sends alerts to contacts who need to be notified. The alerts are promptly sent via email or SMS to the user's family and designated public authorities.

[0050] Step 6:

[0051] The device, upon receiving an alarm, will provide the user with audio and visual warnings. This notification will inform the user of the potential for fraud and encourage them to take appropriate action.

[0052] These steps ensure the system functions in real time to protect users and prevent fraudulent activity.

[0053] (Example 1)

[0054] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0055] In modern society, individuals, including the elderly, are increasingly at risk of becoming victims of fraud. Fraud using voice, in particular, is easy to carry out and prone to causing harm. There is a need for real-time and reliable monitoring and notification methods to protect individuals from such fraud.

[0056] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0057] In this invention, the server includes means for collecting acoustic data, means for utilizing speech recognition technology to convert the acoustic data into text information, and means for analyzing the converted text information and identifying harmful acts using machine learning algorithms. This makes it possible to analyze user conversations in real time, quickly detect the risk of fraud, and appropriately send necessary notifications.

[0058] A "user" refers to a person whose voice data is monitored using the system.

[0059] A "terminal" refers to a device capable of collecting and analyzing acoustic data.

[0060] "Audio data" refers to information about audio waveforms collected by a device.

[0061] "Speech recognition technology" refers to a technical method for converting audio data into textual information.

[0062] "Textual information" refers to data in text format obtained using speech recognition technology.

[0063] A "machine learning algorithm" refers to a programmatic method for analyzing textual information and identifying harmful behavior.

[0064] An "alert" refers to a notification issued when a potential fraud is detected.

[0065] An "emergency contact" refers to a pre-registered individual or organization to which an alert is sent.

[0066] A "public service agency" refers to an organization formed to provide social safety and support.

[0067] This invention provides a system that enables early detection and notification of harmful acts using voice data. This system is based on three components: a terminal, a server, and a user.

[0068] The terminal is a device designed to collect audio from the user's surroundings. It has a built-in microphone system and constantly monitors the user's audio environment. The collected audio data is immediately converted into text using speech recognition software. To achieve high accuracy in this process, speech recognition technologies such as Google® Speech-to-Text API are used.

[0069] The server receives text information sent from the terminal and analyzes it using machine learning algorithms. Specifically, it utilizes generative AI models such as BERT and GPT models to detect potential harmful activities in the text data in real time. The server compares this information with a database of known harmful activities and promptly sends out a notification if fraud is suspected. The notification is sent through an alarm system, and the information is provided to emergency contacts and public service agencies.

[0070] This system allows users to take swift action when suspicious audio is detected. For example, if a user receives a suspicious phone call and it is determined to be a fraudulent billing attempt, the system will immediately alert family members and relevant parties, allowing the user to take appropriate action based on that alert.

[0071] An example of a prompt message would be, "Please describe the features of the fraud detection system that utilizes speech recognition technology." In this way, the present invention provides real-time detection and countermeasures for harmful acts by utilizing voice data.

[0072] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0073] Step 1:

[0074] The device collects audio data from the user's surroundings. Audio data input is obtained through a built-in microphone system. This audio data is recorded as a raw acoustic signal and then formatted into a specific format to facilitate processing in the next step. Furthermore, the recorded audio data is improved in quality through a noise reduction filter.

[0075] Step 2:

[0076] The device converts the collected audio data into text information. This stage utilizes speech recognition technologies such as the Google Speech-to-Text API. It receives audio data as input and generates text information (text data) as output. This conversion process involves analyzing the characteristics of the audio signal and selecting appropriate words. Finally, the audio conversation is output in a text format that is easy to understand.

[0077] Step 3:

[0078] The server processes text information received from the terminal. It takes text data as input and uses machine learning algorithms to detect potential harmful activities within it. Specifically, it uses generative AI models such as BERT and GPT to analyze specific phrases and contexts within the text. The output provides an assessment of the likelihood of fraud or harmful activity. This information serves as foundational data for the next step.

[0079] Step 4:

[0080] The server issues an alert if it determines that a scam is highly likely. Here, it uses the evaluation results obtained from step 3 as input to generate notifications for family members and public service agencies. The notifications include details of the fraudulent activity and the user's location information to facilitate a quick response to the emergency. As output, real-time alerts are sent.

[0081] Step 5:

[0082] The user receives an alert from the system and immediately recognizes the risk. Upon receiving the notification, the user can take appropriate action based on the information provided. At this stage, it is recommended that the user's family and related parties also cooperate to take action to protect the user from potential dangers.

[0083] (Application Example 1)

[0084] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0085] Currently, the number of cases in which users, including the elderly, fall victim to fraud is increasing, making fraud prevention a critical social issue. Traditional methods often result in delays in detecting fraud and insufficient response. Therefore, there is a need for a system that monitors the audio environment in real time and quickly detects and notifies users of potential fraud.

[0086] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0087] In this invention, the server includes acoustic processing means for monitoring the user's voice environment and collecting acoustic data, data conversion means for converting the collected acoustic data into text information, and analysis means for comparing the converted text information with existing fraud patterns to determine the likelihood of fraud. This makes it possible to monitor the likelihood of fraud in real time and issue warnings quickly.

[0088] "Acoustic processing means" refers to a device that continuously monitors the sound environment surrounding the user and collects necessary acoustic data.

[0089] A "data conversion means" is a device that converts collected acoustic data into textual information and generates text using high-precision speech recognition technology.

[0090] The "analysis tool" is a device that compares the converted text information with existing fraud patterns to determine the likelihood of fraud, and performs real-time analysis based on AI technology.

[0091] A "notification system" is a device that, when a potential scam is detected, issues a warning to pre-registered contacts, prompting them to take prompt action.

[0092] "Portable information devices" refer to portable information terminals such as smartphones and smart glasses, which serve as platforms that integrate sound processing capabilities.

[0093] To implement this invention, it is necessary to construct a system that mainly includes acoustic processing means, data conversion means, analysis means, and notification means. It is desirable that the system be integrated into a portable information device such as a smartphone or smart glasses.

[0094] The device uses acoustic processing to monitor audio data around the user in real time. This acoustic data is converted into highly accurate text information by a data conversion tool using the Google Cloud Speech-to-Text API. The converted text information is analyzed on a server using AI technologies such as TENSORFLOW (registered trademark) and compared with existing fraud patterns to determine the likelihood of fraud.

[0095] If a user is deemed highly likely to be a scam, the server uses notification methods such as the Twilio API to send a warning to pre-registered contacts. This allows the user's family or caregivers to take immediate action.

[0096] For example, if a user receives a phone call from someone asking for their bank information, the device collects and analyzes the audio. If it determines that the call may be fraudulent, it immediately sends an alert to family members saying, "A suspicious call has been detected. Please check."

[0097] An example of a prompt message that would be effective is: "Please analyze the text of this audio and assess the likelihood of fraud. List any suspicious phrases or contexts and notify me of the results."

[0098] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0099] Step 1:

[0100] The terminal acquires audio data from the user's surroundings using acoustic processing equipment. The input is an audio signal, and the output is audio data. At this stage, the operation involves capturing the audio signal through the microphone.

[0101] Step 2:

[0102] The device converts the collected audio data into text using a data conversion method. The input is audio data, and the output is text data. This conversion uses the Google Cloud Speech-to-Text API to perform highly accurate speech recognition.

[0103] Step 3:

[0104] The server analyzes the text data received from the data transformation device using an analysis device. The input is text data, and the output is an evaluation result of whether or not it is potentially fraudulent. Specifically, it uses an AI model based on TensorFlow to compare the text data with existing fraud patterns.

[0105] Step 4:

[0106] If the server determines, based on the analysis results, that a scam is highly likely, it will send an alert to pre-registered contacts using a notification method. The input is the scam assessment result, and the output is a warning message. Here, the process of sending emails or SMS messages is performed using the Twilio API, etc.

[0107] Step 5:

[0108] The user's family and care staff receive warning messages from the device and take prompt action. The input is the warning message, and the output is the actual response action. Specific actions include contacting or visiting the user to ensure their safety.

[0109] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0110] This invention relates to a system that enables a user to quickly detect situations in which they are at risk of fraud and to take appropriate action. This system includes an emotion engine that has the capability to recognize the user's emotions.

[0111] The emotion engine is designed to analyze the user's emotions in real time from their voice. The device uses its acoustic processing unit to monitor the user's conversation and collect voice data, while the emotion engine extracts emotional data. The extracted emotional data is used to capture the user's stress level and signs of anxiety.

[0112] Once voice data is collected, the terminal converts it into text data using a data conversion unit. In parallel, the emotion engine records changes in emotional state and passes this data to a comparison device. The server uses the received text data and emotion data to compare and analyze them with existing fraud patterns and assess the likelihood of fraud.

[0113] The server comprehensively assesses the likelihood of fraud and the user's emotional state, and the alert system sets the alert level. Based on this assessment, the alert system sends notifications to the user's family and public institutions. The urgency of the alert is adjusted based on emotional data, so if there is a significant change in the user's emotions, immediate action is prompted.

[0114] For example, if a user comes into contact with a scammer over the phone, and their anxiety increases during the conversation, the emotion engine will immediately detect this change. This allows the system to add emotional information to its usual scam pattern detection and quickly issue alerts, enabling a faster response.

[0115] In this way, this system, equipped with an emotion engine, further enhances user safety by detecting potential fraud with high accuracy and issuing timely warnings.

[0116] The following describes the processing flow.

[0117] Step 1:

[0118] The device uses a microphone to monitor the user's conversation in real time and collect audio data. An acoustic processing unit reduces noise and ensures a clear audio signal.

[0119] Step 2:

[0120] The device inputs the collected audio data into an emotion engine, which analyzes parameters such as tone, pitch, and speed to identify the user's emotional state. In particular, it looks for signs of anxiety or stress.

[0121] Step 3:

[0122] The terminal simultaneously sends audio data to the data conversion unit, which converts the audio to text. Speech recognition technology is used to accurately transcribe the conversation into text.

[0123] Step 4:

[0124] The server receives emotional data from the emotion engine and text data from the data transformation unit. After receiving the data, it applies an AI algorithm to compare the emotional changes with known patterns of fraud.

[0125] Step 5:

[0126] The server assesses the risk of fraud and increases the risk level if the user is under high stress. It makes a comprehensive judgment on the likelihood of fraudulent activity.

[0127] Step 6:

[0128] The server, via its alarm system, sends alerts to the user's family and public institutions based on the detected fraud risk. The urgency of the alert is adjusted based on the degree of emotional change.

[0129] Step 7:

[0130] The device receives an alert and provides the user with a visual or auditory warning. This allows the user to recognize the situation and take appropriate action.

[0131] This process allows the system to take the user's emotional state into account and enhance its vigilance against fraud.

[0132] (Example 2)

[0133] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0134] Conventional fraud detection systems fail to take into account the user's emotional state, making it difficult to quickly and accurately assess the likelihood of fraud. Furthermore, there is a lack of information to enable users to take appropriate action in emergency situations when they are exposed to fraud risks.

[0135] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0136] In this invention, the server includes means for collecting acoustic signals and extracting emotional information, means for converting acoustic signals into textual information, and means for evaluating the possibility of fraud using the textual information and emotional information. This makes it possible to accurately grasp changes in the user's emotions, quickly assess the risk of fraud, and notify external organizations.

[0137] A "user" refers to an individual or organization that utilizes a system.

[0138] An "acoustic signal" is audio data generated by a user's conversation, and is an analog or digital sound waveform collected by a device such as a microphone.

[0139] "Device means" refers to physical or virtual devices or equipment used to achieve a specific function.

[0140] "Emotional information" refers to data extracted from acoustic signals, and includes indicators and categories that show the user's emotional state.

[0141] "Textual information" refers to text data obtained by converting acoustic signals, and is a string of characters that represents the content of a conversation.

[0142] A "fraud pattern" is a dataset that compiles the characteristics and methods of fraudulent activities reported in the past, and is used to evaluate the potential of new frauds.

[0143] "External organization" refers to an external group, individual, or public institution to which warnings or notifications are sent.

[0144] The embodiments for carrying out the present invention will be described below.

[0145] The device includes an acoustic processing unit for monitoring user conversations in real time. This unit uses a microphone to collect acoustic signals and converts them into digital signals. This digitized audio data is sent to an emotion engine to extract the user's emotions. The emotion engine uses machine learning algorithms to determine emotional information from the acoustic signals in real time. It is also possible to utilize external audio processing platforms (such as Amazon Web Services or Google Cloud Platform's speech analysis APIs) for this process.

[0146] The collected audio data is converted into text information via speech recognition software. For this conversion, for example, the speech recognition API of Microsoft® Azure® can be used.

[0147] The server uses textual and sentimental information received from the terminal to compare and analyze it against existing fraud patterns. This process utilizes generative AI models and natural language processing techniques to assess the likelihood of fraud. Based on the evaluated data, the server sets a warning level and sends necessary notifications to external organizations. In this process, it is conceivable that warnings would be sent via SMS or email using communication APIs such as Twilio.

[0148] For example, if a user receives a potentially fraudulent phone call, the device detects changes in the user's voice during the conversation, and the emotion engine immediately detects an increase in anxiety. The server analyzes the emotional changes and text information, and if it assesses a high probability of fraud, it sets an "urgent" alert level and sends a notification to the user's family and relevant organizations.

[0149] Examples of input prompts for a generative AI model:

[0150] "How does the emotion engine respond when a user receives a potentially fraudulent phone call? Please describe in detail the system's response based on emotion detection."

[0151] In this way, the system enables highly accurate fraud detection based on changes in the user's emotions and enhances user safety through timely alarm notifications.

[0152] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0153] Step 1:

[0154] The terminal collects acoustic signals in real time using a microphone to receive user conversations. The input for this step is ambient noise and user speech. The terminal converts the acoustic signals into digital data and performs signal processing such as noise filtering to clearly extract the user's voice. The output is clear digital audio data.

[0155] Step 2:

[0156] The device sends the collected digital audio data to the emotion engine, which then extracts emotional information. The input for this step is framed audio data. The device analyzes features such as tone, volume changes, and speed of the audio and uses a machine learning algorithm to classify the user's emotional state. The output is data indicating the user's emotion (e.g., "relieved," "tense," "anxious").

[0157] Step 3:

[0158] The device uses speech recognition software to convert digital audio data into text. The input for this step is pre-processed digital audio data. The device converts the audio to text, for example, using the Google Speech-to-Text API. The resulting text data represents the conversation content as written text.

[0159] Step 4:

[0160] The server analyzes the likelihood of fraud using text and sentiment information received from the terminal. The input for this step is text data and sentiment data. The server uses natural language processing techniques to analyze the text information and compare it to existing fraud patterns in the database. It also uses a generative AI model to predict unknown fraud patterns. The output is evaluation data indicating the likelihood and confidence level of fraud.

[0161] Step 5:

[0162] The server sets the alert level and sends out notifications. The input for this step is fraud rating data and user sentiment data. Based on the rating data, the server determines an alert level such as "Low," "Medium," "High," or "Emergency." Based on this, it transmits alerts to family members or public institutions using communication APIs such as Twilio. The output is the set alert level and a message notifying them of this.

[0163] Each of these processing steps allows the system to capture changes in the user's emotions while efficiently detecting and notifying them of potential fraud.

[0164] (Application Example 2)

[0165] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0166] Conventional fraud detection systems rely solely on comparing voice data and fail to consider changes in the user's emotions, making it difficult to accurately determine the likelihood of fraud. Furthermore, they may not adequately reflect the user's anxiety or stress levels, potentially leading to delays in issuing timely warnings.

[0167] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0168] In this invention, the server includes data conversion means for converting voice information into text, data comparison means for comparing the converted text with fraud patterns, emotion analysis means for analyzing emotions from voice in real time, and alarm issuing means for issuing notifications based on the likelihood of fraud and emotion analysis. This enables highly accurate fraud detection and timely alarm issuance that takes into account the user's emotional state.

[0169] A "user" is an individual or legal entity that uses the system to make calls or communicate.

[0170] A "telephone call" is a real-time communication method that uses voice to exchange information.

[0171] "Audio information" refers to audio data that includes the words and utterances spoken by the user.

[0172] "Acoustic analysis means" refers to a device or system for acquiring audio information and performing necessary processing.

[0173] "Data conversion means" refers to a device or program that has the function of converting audio information into text data.

[0174] A "data comparison means" is a device or program that compares converted text with existing patterns and evaluates the degree of matching or similarity.

[0175] An "emotion analysis device" is a device or system that analyzes a user's emotions in real time from their voice and evaluates the degree of stress and anxiety.

[0176] A "fraud pattern" is data that shows typical combinations of behaviors and expressions based on past fraudulent activities.

[0177] An "alarm notification device" is a device or system that has the function of notifying in a pre-set manner when it detects the possibility of fraud.

[0178] A "portable communication device" is a portable device that has communication capabilities for sending and receiving voice and data while on the move.

[0179] A "server" is a computer system or device that processes data over a network and provides services to clients.

[0180] The system for realizing this invention includes acoustic analysis means, data conversion means, data comparison means, emotion analysis means, and alarm issuing means. This system enables faster and more accurate detection of potential fraud by reflecting the user's emotional state in real time in conventional fraud detection technology.

[0181] The acoustic analysis system uses the microphone of a mobile communication device to collect user voice information. This collection process uses speech analysis software, such as the Google Cloud Speech-to-Text API, to convert the voice data into text data. The converted text data is then sent to a server.

[0182] The server first uses a data comparison tool to match text data against existing fraud patterns. Next, an emotion analysis tool analyzes the user's emotions extracted from the audio information and evaluates their stress and anxiety levels. This analysis utilizes emotion analysis models based on TensorFlow or PyTorch.

[0183] After integrating the user's emotional state with the results of data comparison, the server uses an alarm system to issue a warning if fraud is suspected. This warning is sent to the user themselves, or, in some cases, to appropriate contacts such as family members or public institutions. The urgency of the notification is flexibly adjusted according to the magnitude of the user's emotional change, enabling a quick response.

[0184] For example, if an unstable or high-stress state is detected when a user provides credit card information in an e-commerce transaction, an alert will be issued immediately. This operation strengthens defenses against fraudulent activity and improves user safety.

[0185] An example of a prompt might be: "Design a voice sentiment analysis app that detects potential fraud during a call. Generate a prototype that issues a warning when the user becomes anxious, and explain how the system works."

[0186] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0187] Step 1:

[0188] The terminal collects user voice information in real time using the microphone of the mobile communication device. The input is raw voice data, which is transmitted directly to the acoustic analysis function. Using the acoustic analysis means, noise in the voice information is removed and output as clear voice data.

[0189] Step 2:

[0190] The terminal inputs clear audio data obtained by the acoustic analysis means into the data conversion means. Here, the Google Cloud Speech-to-Text API is used to convert the audio data into text data. The output is the converted text data.

[0191] Step 3:

[0192] The server inputs the text data sent from the terminal into a data comparison device. Here, the text data is compared to predefined fraud patterns, and a match or similarity is evaluated. A fraud probability score is generated as output.

[0193] Step 4:

[0194] The terminal then again supplies the user's voice information in real time, this time to the emotion analysis system. Emotional data is extracted from the voice using an emotion analysis model based on TensorFlow or PyTorch. The input is raw voice data, and the output is the user's emotional state (e.g., stress level, degree of anxiety).

[0195] Step 5:

[0196] The server integrates the fraud probability score from Step 3 and the emotional state from Step 4 to make an overall judgment. Here, it performs data calculations to determine the urgency of a fraud alert. The output is notification data that includes the urgency for issuing an alert.

[0197] Step 6:

[0198] The server issues notifications based on the generated urgency level through an alarm system. The input is notification data, and the output is an alarm sent to the user or a designated contact (e.g., family, public facilities). The content of the alarm is customized according to the likelihood of fraud and changes in the user's emotional state.

[0199] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0200] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0201] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0202] [Second Embodiment]

[0203] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0204] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0205] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0206] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0207] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0208] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0209] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0210] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0211] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0212] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0213] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0214] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0215] This invention relates to an audio data monitoring system for protecting users, including the elderly, from fraud. The system consists of an acoustic processing unit, a data conversion unit, a comparison unit, and an alarm issuing unit.

[0216] First, the device constantly monitors the sounds around the user and uses an acoustic processing unit to collect conversations. Once the audio data is collected, a data conversion unit within the system receives it and converts it from audio to text data. During this conversion process, speech recognition technology generates highly accurate text.

[0217] Next, the server retrieves the converted text data and compares it to a database of previously collected fraudulent activity. This process utilizes AI algorithms to analyze and identify potential fraud in real time. In particular, it is designed to effectively detect specific phrases and contexts that may be associated with fraud.

[0218] If a fraudulent activity is deemed highly likely, the server controls the alarm system to send an alert to the user's family and pre-registered public institutions. This alert includes details about the potential fraud and the user's current location, allowing for a swift response.

[0219] As a concrete example of implementation, consider a scenario where a user receives a suspicious phone call and it is identified as a fraudulent billing attempt. In this case, the device monitors the conversation, processes the necessary information, and immediately warns the family, allowing the user to take steps to prevent becoming a victim of fraud.

[0220] In this way, this system can dynamically and quickly detect fraud using voice information and send alerts to relevant parties, thereby preventing fraud from occurring.

[0221] The following describes the processing flow.

[0222] Step 1:

[0223] The device monitors the sounds around the user in real time and collects them as audio data. Using an acoustic processing device, it reduces ambient noise while clearly capturing the content of conversations.

[0224] Step 2:

[0225] The terminal transmits the collected audio data to the data conversion unit. Here, the audio data is converted into text data using speech recognition software. The converted text is processed to accurately represent the conversation content without being affected by pronunciation quirks or noise.

[0226] Step 3:

[0227] The server receives the text data and applies an AI algorithm to compare it with a fraud database. The AI ​​uses machine learning models to identify unknown tactics as well as matching against known fraud patterns. At this stage, the likelihood of fraud is analyzed.

[0228] Step 4:

[0229] If the server determines that there is a high risk of fraud, it sends information to the alerting unit. The notification is then generated to include details of the suspected fraud and the user's location information.

[0230] Step 5:

[0231] The server sends alerts to contacts who need to be notified. The alerts are promptly sent via email or SMS to the user's family and designated public authorities.

[0232] Step 6:

[0233] The device, upon receiving an alarm, will provide the user with audio and visual warnings. This notification will inform the user of the potential for fraud and encourage them to take appropriate action.

[0234] These steps ensure the system functions in real time to protect users and prevent fraudulent activity.

[0235] (Example 1)

[0236] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0237] In modern society, individuals, including the elderly, are increasingly at risk of becoming victims of fraud. Fraud using voice, in particular, is easy to carry out and prone to causing harm. There is a need for real-time and reliable monitoring and notification methods to protect individuals from such fraud.

[0238] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0239] In this invention, the server includes means for collecting acoustic data, means for utilizing speech recognition technology to convert the acoustic data into text information, and means for analyzing the converted text information and identifying harmful acts using machine learning algorithms. This makes it possible to analyze user conversations in real time, quickly detect the risk of fraud, and appropriately send necessary notifications.

[0240] A "user" refers to a person whose voice data is monitored using the system.

[0241] A "terminal" refers to a device capable of collecting and analyzing acoustic data.

[0242] "Audio data" refers to information about audio waveforms collected by a device.

[0243] "Speech recognition technology" refers to a technical method for converting audio data into textual information.

[0244] "Textual information" refers to data in text format obtained using speech recognition technology.

[0245] A "machine learning algorithm" refers to a programmatic method for analyzing textual information and identifying harmful behavior.

[0246] An "alert" refers to a notification issued when a potential fraud is detected.

[0247] An "emergency contact" refers to a pre-registered individual or organization to which an alert is sent.

[0248] A "public service agency" refers to an organization formed to provide social safety and support.

[0249] This invention provides a system that enables early detection and notification of harmful acts using voice data. This system is based on three components: a terminal, a server, and a user.

[0250] The device is designed to collect audio from the user's surroundings. It incorporates a microphone system and constantly monitors the user's audio environment. The collected audio data is immediately converted into text using speech recognition software. To achieve high accuracy in this process, speech recognition technologies such as the Google Speech-to-Text API are used.

[0251] The server receives text information sent from the terminal and analyzes it using machine learning algorithms. Specifically, it utilizes generative AI models such as BERT and GPT models to detect potential harmful activities in the text data in real time. The server compares this information with a database of known harmful activities and promptly sends out a notification if fraud is suspected. The notification is sent through an alarm system, and the information is provided to emergency contacts and public service agencies.

[0252] This system allows users to take swift action when suspicious audio is detected. For example, if a user receives a suspicious phone call and it is determined to be a fraudulent billing attempt, the system will immediately alert family members and relevant parties, allowing the user to take appropriate action based on that alert.

[0253] An example of a prompt message would be, "Please describe the features of the fraud detection system that utilizes speech recognition technology." In this way, the present invention provides real-time detection and countermeasures for harmful acts by utilizing voice data.

[0254] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0255] Step 1:

[0256] The device collects audio data from the user's surroundings. Audio data input is obtained through a built-in microphone system. This audio data is recorded as a raw acoustic signal and then formatted into a specific format to facilitate processing in the next step. Furthermore, the recorded audio data is improved in quality through a noise reduction filter.

[0257] Step 2:

[0258] The device converts the collected audio data into text information. This stage utilizes speech recognition technologies such as the Google Speech-to-Text API. It receives audio data as input and generates text information (text data) as output. This conversion process involves analyzing the characteristics of the audio signal and selecting appropriate words. Finally, the audio conversation is output in a text format that is easy to understand.

[0259] Step 3:

[0260] The server processes text information received from the terminal. It takes text data as input and uses machine learning algorithms to detect potential harmful activities within it. Specifically, it uses generative AI models such as BERT and GPT to analyze specific phrases and contexts within the text. The output provides an assessment of the likelihood of fraud or harmful activity. This information serves as foundational data for the next step.

[0261] Step 4:

[0262] The server issues an alert if it determines that a scam is highly likely. Here, it uses the evaluation results obtained from step 3 as input to generate notifications for family members and public service agencies. The notifications include details of the fraudulent activity and the user's location information to facilitate a quick response to the emergency. As output, real-time alerts are sent.

[0263] Step 5:

[0264] The user receives an alert from the system and immediately recognizes the risk. Upon receiving the notification, the user can take appropriate action based on the information provided. At this stage, it is recommended that the user's family and related parties also cooperate to take action to protect the user from potential dangers.

[0265] (Application Example 1)

[0266] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0267] Currently, the number of cases in which users, including the elderly, fall victim to fraud is increasing, making fraud prevention a critical social issue. Traditional methods often result in delays in detecting fraud and insufficient response. Therefore, there is a need for a system that monitors the audio environment in real time and quickly detects and notifies users of potential fraud.

[0268] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0269] In this invention, the server includes acoustic processing means for monitoring the user's voice environment and collecting acoustic data, data conversion means for converting the collected acoustic data into text information, and analysis means for comparing the converted text information with existing fraud patterns to determine the likelihood of fraud. This makes it possible to monitor the likelihood of fraud in real time and issue warnings quickly.

[0270] "Acoustic processing means" refers to a device that continuously monitors the sound environment surrounding the user and collects necessary acoustic data.

[0271] A "data conversion means" is a device that converts collected acoustic data into textual information and generates text using high-precision speech recognition technology.

[0272] The "analysis tool" is a device that compares the converted text information with existing fraud patterns to determine the likelihood of fraud, and performs real-time analysis based on AI technology.

[0273] A "notification system" is a device that, when a potential scam is detected, issues a warning to pre-registered contacts, prompting them to take prompt action.

[0274] "Portable information devices" refer to portable information terminals such as smartphones and smart glasses, which serve as platforms that integrate sound processing capabilities.

[0275] To implement this invention, it is necessary to construct a system that mainly includes acoustic processing means, data conversion means, analysis means, and notification means. It is desirable that the system be integrated into a portable information device such as a smartphone or smart glasses.

[0276] The device uses acoustic processing to monitor audio data around the user in real time. This acoustic data is converted into highly accurate text information by a data conversion tool using the Google Cloud Speech-to-Text API. The converted text information is analyzed on a server using AI technologies such as TensorFlow, and the likelihood of fraud is determined by comparing it with existing fraud patterns.

[0277] If a user is deemed highly likely to be a scam, the server uses notification methods such as the Twilio API to send a warning to pre-registered contacts. This allows the user's family or caregivers to take immediate action.

[0278] For example, if a user receives a phone call from someone asking for their bank information, the device collects and analyzes the audio. If it determines that the call may be fraudulent, it immediately sends an alert to family members saying, "A suspicious call has been detected. Please check."

[0279] An example of a prompt message that would be effective is: "Please analyze the text of this audio and assess the likelihood of fraud. List any suspicious phrases or contexts and notify me of the results."

[0280] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0281] Step 1:

[0282] The terminal acquires audio data from the user's surroundings using acoustic processing equipment. The input is an audio signal, and the output is audio data. At this stage, the operation involves capturing the audio signal through the microphone.

[0283] Step 2:

[0284] The terminal converts the collected voice data into text using data conversion means. The input is voice data, and the output is text data. Google Cloud Speech-to-Text API is used for this conversion to perform high-precision speech recognition.

[0285] Step 3:

[0286] The server analyzes the text data received from the data conversion means by means of analysis means. The input is text data, and the output is an evaluation result as to whether there is a possibility of fraud. As a specific operation, an AI model using TensorFlow is used to compare the text data with existing fraud patterns.

[0287] Step 4:

[0288] If the server determines based on the analysis result that there is a high possibility of fraud, it sends an alert to a pre-registered contact using notification means. The input is the fraud evaluation result, and the output is a warning message. Here, operations such as sending emails and SMS are performed using the Twilio API or the like.

[0289] Step 5:

[0290] The user's family members and care staff receive the warning message from the terminal and take prompt action. The input is the warning message, and the output is the actual action taken. As a specific operation, they contact or visit the user to confirm safety.

[0291] Furthermore, an emotion engine for estimating the user's emotion may be combined. That is, the specific processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform specific processing using the user's emotion.

[0292] This invention relates to a system that enables a user to quickly detect situations in which they are at risk of fraud and to take appropriate action. This system includes an emotion engine that has the capability to recognize the user's emotions.

[0293] The emotion engine is designed to analyze the user's emotions in real time from their voice. The device uses its acoustic processing unit to monitor the user's conversation and collect voice data, while the emotion engine extracts emotional data. The extracted emotional data is used to capture the user's stress level and signs of anxiety.

[0294] Once voice data is collected, the terminal converts it into text data using a data conversion unit. In parallel, the emotion engine records changes in emotional state and passes this data to a comparison device. The server uses the received text data and emotion data to compare and analyze them with existing fraud patterns and assess the likelihood of fraud.

[0295] The server comprehensively assesses the likelihood of fraud and the user's emotional state, and the alert system sets the alert level. Based on this assessment, the alert system sends notifications to the user's family and public institutions. The urgency of the alert is adjusted based on emotional data, so if there is a significant change in the user's emotions, immediate action is prompted.

[0296] For example, if a user comes into contact with a scammer over the phone, and their anxiety increases during the conversation, the emotion engine will immediately detect this change. This allows the system to add emotional information to its usual scam pattern detection and quickly issue alerts, enabling a faster response.

[0297] In this way, this system, equipped with an emotion engine, further enhances user safety by detecting potential fraud with high accuracy and issuing timely warnings.

[0298] The following describes the processing flow.

[0299] Step 1:

[0300] The terminal uses a microphone to monitor the user's conversation in real time and collect voice data. The acoustic processing device reduces noise and ensures a clear voice signal.

[0301] Step 2:

[0302] The terminal inputs the collected voice data into an emotion engine, analyzes parameters such as the tone, pitch, and speed of the voice to identify the user's emotional state. At this time, signs indicating anxiety or stress are particularly sought.

[0303] Step 3:

[0304] The terminal simultaneously sends the voice data to a data conversion unit for conversion from voice to text. The accurate conversation content is texturized using speech recognition technology.

[0305] Step 4:

[0306] The server receives the emotion data from the emotion engine and the text data from the data conversion unit. After receiving, an AI algorithm is applied to compare the change in emotion with known patterns of fraud.

[0307] Step 5:

[0308] The server evaluates the risk of fraud and raises the risk level if the user is in a high-stress state. The possibility of fraud is comprehensively judged.

[0309] Step 6:

[0310] The server transmits an alert corresponding to the detected fraud risk to the user's family or public institutions through an alert transmission unit. The urgency of the alert is adjusted based on the degree of emotional change.

[0311] Step 7:

[0312] The device receives an alert and provides the user with a visual or auditory warning. This allows the user to recognize the situation and take appropriate action.

[0313] This process allows the system to take the user's emotional state into account and enhance its vigilance against fraud.

[0314] (Example 2)

[0315] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0316] Conventional fraud detection systems fail to take into account the user's emotional state, making it difficult to quickly and accurately assess the likelihood of fraud. Furthermore, there is a lack of information to enable users to take appropriate action in emergency situations when they are exposed to fraud risks.

[0317] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0318] In this invention, the server includes means for collecting acoustic signals and extracting emotional information, means for converting acoustic signals into textual information, and means for evaluating the possibility of fraud using the textual information and emotional information. This makes it possible to accurately grasp changes in the user's emotions, quickly assess the risk of fraud, and notify external organizations.

[0319] A "user" refers to an individual or organization that utilizes a system.

[0320] An "acoustic signal" is audio data generated by a user's conversation, and is an analog or digital sound waveform collected by a device such as a microphone.

[0321] "Device means" refers to physical or virtual devices or equipment used to achieve a specific function.

[0322] "Emotional information" refers to data extracted from acoustic signals, and includes indicators and categories that show the user's emotional state.

[0323] "Textual information" refers to text data obtained by converting acoustic signals, and is a string of characters that represents the content of a conversation.

[0324] A "fraud pattern" is a dataset that compiles the characteristics and methods of fraudulent activities reported in the past, and is used to evaluate the potential of new frauds.

[0325] "External organization" refers to an external group, individual, or public institution to which warnings or notifications are sent.

[0326] The embodiments for carrying out the present invention will be described below.

[0327] The device includes an acoustic processing unit for monitoring user conversations in real time. This unit uses a microphone to collect acoustic signals and converts them into digital signals. This digitized audio data is sent to an emotion engine to extract the user's emotions. The emotion engine uses machine learning algorithms to determine emotional information from the acoustic signals in real time. It is also possible to utilize external audio processing platforms (such as Amazon Web Services or Google Cloud Platform's speech analysis APIs) for this process.

[0328] The collected audio data is converted into text information via speech recognition software. For example, the Microsoft Azure speech recognition API can be used for this conversion.

[0329] The server uses textual and sentimental information received from the terminal to compare and analyze it against existing fraud patterns. This process utilizes generative AI models and natural language processing techniques to assess the likelihood of fraud. Based on the evaluated data, the server sets a warning level and sends necessary notifications to external organizations. In this process, it is conceivable that warnings would be sent via SMS or email using communication APIs such as Twilio.

[0330] For example, if a user receives a potentially fraudulent phone call, the device detects changes in the user's voice during the conversation, and the emotion engine immediately detects an increase in anxiety. The server analyzes the emotional changes and text information, and if it assesses a high probability of fraud, it sets an "urgent" alert level and sends a notification to the user's family and relevant organizations.

[0331] Examples of input prompts for a generative AI model:

[0332] "How does the emotion engine respond when a user receives a potentially fraudulent phone call? Please describe in detail the system's response based on emotion detection."

[0333] In this way, the system enables highly accurate fraud detection based on changes in the user's emotions and enhances user safety through timely alarm notifications.

[0334] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0335] Step 1:

[0336] The terminal collects acoustic signals in real time using a microphone to receive user conversations. The input for this step is ambient noise and user speech. The terminal converts the acoustic signals into digital data and performs signal processing such as noise filtering to clearly extract the user's voice. The output is clear digital audio data.

[0337] Step 2:

[0338] The device sends the collected digital audio data to the emotion engine, which then extracts emotional information. The input for this step is framed audio data. The device analyzes features such as tone, volume changes, and speed of the audio and uses a machine learning algorithm to classify the user's emotional state. The output is data indicating the user's emotion (e.g., "relieved," "tense," "anxious").

[0339] Step 3:

[0340] The device uses speech recognition software to convert digital audio data into text. The input for this step is pre-processed digital audio data. The device converts the audio to text, for example, using the Google Speech-to-Text API. The resulting text data represents the conversation content as written text.

[0341] Step 4:

[0342] The server analyzes the likelihood of fraud using text and sentiment information received from the terminal. The input for this step is text data and sentiment data. The server uses natural language processing techniques to analyze the text information and compare it to existing fraud patterns in the database. It also uses a generative AI model to predict unknown fraud patterns. The output is evaluation data indicating the likelihood and confidence level of fraud.

[0343] Step 5:

[0344] The server sets the alert level and sends out notifications. The input for this step is fraud rating data and user sentiment data. Based on the rating data, the server determines an alert level such as "Low," "Medium," "High," or "Emergency." Based on this, it transmits alerts to family members or public institutions using communication APIs such as Twilio. The output is the set alert level and a message notifying them of this.

[0345] Each of these processing steps allows the system to capture changes in the user's emotions while efficiently detecting and notifying them of potential fraud.

[0346] (Application Example 2)

[0347] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[0348] Conventional fraud detection systems rely solely on comparing voice data and fail to consider changes in the user's emotions, making it difficult to accurately determine the likelihood of fraud. Furthermore, they may not adequately reflect the user's anxiety or stress levels, potentially leading to delays in issuing timely warnings.

[0349] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0350] In this invention, the server includes data conversion means for converting voice information into text, data comparison means for comparing the converted text with fraud patterns, emotion analysis means for analyzing emotions from voice in real time, and alarm issuing means for issuing notifications based on the likelihood of fraud and emotion analysis. This enables highly accurate fraud detection and timely alarm issuance that takes into account the user's emotional state.

[0351] A "user" is an individual or legal entity that uses the system to make calls or communicate.

[0352] A "telephone call" is a real-time communication method that uses voice to exchange information.

[0353] "Audio information" refers to audio data that includes the words and utterances spoken by the user.

[0354] "Acoustic analysis means" refers to a device or system for acquiring audio information and performing necessary processing.

[0355] "Data conversion means" refers to a device or program that has the function of converting audio information into text data.

[0356] A "data comparison means" is a device or program that compares converted text with existing patterns and evaluates the degree of matching or similarity.

[0357] An "emotion analysis device" is a device or system that analyzes a user's emotions in real time from their voice and evaluates the degree of stress and anxiety.

[0358] A "fraud pattern" is data that shows typical combinations of behaviors and expressions based on past fraudulent activities.

[0359] An "alarm notification device" is a device or system that has the function of notifying in a pre-set manner when it detects the possibility of fraud.

[0360] A "portable communication device" is a portable device that has communication capabilities for sending and receiving voice and data while on the move.

[0361] A "server" is a computer system or device that processes data over a network and provides services to clients.

[0362] The system for realizing this invention includes acoustic analysis means, data conversion means, data comparison means, emotion analysis means, and alarm issuing means. This system enables faster and more accurate detection of potential fraud by reflecting the user's emotional state in real time in conventional fraud detection technology.

[0363] The acoustic analysis system uses the microphone of a mobile communication device to collect user voice information. This collection process uses speech analysis software, such as the Google Cloud Speech-to-Text API, to convert the voice data into text data. The converted text data is then sent to a server.

[0364] The server first uses a data comparison tool to match text data against existing fraud patterns. Next, an emotion analysis tool analyzes the user's emotions extracted from the audio information and evaluates their stress and anxiety levels. This analysis utilizes emotion analysis models based on TensorFlow or PyTorch.

[0365] After integrating the user's emotional state with the results of data comparison, the server uses an alarm system to issue a warning if fraud is suspected. This warning is sent to the user themselves, or, in some cases, to appropriate contacts such as family members or public institutions. The urgency of the notification is flexibly adjusted according to the magnitude of the user's emotional change, enabling a quick response.

[0366] For example, if an unstable or high-stress state is detected when a user provides credit card information in an e-commerce transaction, an alert will be issued immediately. This operation strengthens defenses against fraudulent activity and improves user safety.

[0367] An example of a prompt might be: "Design a voice sentiment analysis app that detects potential fraud during a call. Generate a prototype that issues a warning when the user becomes anxious, and explain how the system works."

[0368] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0369] Step 1:

[0370] The terminal collects user voice information in real time using the microphone of the mobile communication device. The input is raw voice data, which is transmitted directly to the acoustic analysis function. Using the acoustic analysis means, noise in the voice information is removed and output as clear voice data.

[0371] Step 2:

[0372] The terminal inputs clear audio data obtained by the acoustic analysis means into the data conversion means. Here, the Google Cloud Speech-to-Text API is used to convert the audio data into text data. The output is the converted text data.

[0373] Step 3:

[0374] The server inputs the text data sent from the terminal into a data comparison device. Here, the text data is compared to predefined fraud patterns, and a match or similarity is evaluated. A fraud probability score is generated as output.

[0375] Step 4:

[0376] The terminal then again supplies the user's voice information in real time, this time to the emotion analysis system. Emotional data is extracted from the voice using an emotion analysis model based on TensorFlow or PyTorch. The input is raw voice data, and the output is the user's emotional state (e.g., stress level, degree of anxiety).

[0377] Step 5:

[0378] The server integrates the fraud probability score from Step 3 and the emotional state from Step 4 to make an overall judgment. Here, it performs data calculations to determine the urgency of a fraud alert. The output is notification data that includes the urgency for issuing an alert.

[0379] Step 6:

[0380] The server issues notifications based on the generated urgency level through an alarm system. The input is notification data, and the output is an alarm sent to the user or a designated contact (e.g., family, public facilities). The content of the alarm is customized according to the likelihood of fraud and changes in the user's emotional state.

[0381] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0382] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0383] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0384] [Third Embodiment]

[0385] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0386] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0387] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0388] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0389] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0390] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0391] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0392] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0393] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0394] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0395] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0396] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0397] This invention relates to an audio data monitoring system for protecting users, including the elderly, from fraud. The system consists of an acoustic processing unit, a data conversion unit, a comparison unit, and an alarm issuing unit.

[0398] First, the device constantly monitors the sounds around the user and uses an acoustic processing unit to collect conversations. Once the audio data is collected, a data conversion unit within the system receives it and converts it from audio to text data. During this conversion process, speech recognition technology generates highly accurate text.

[0399] Next, the server retrieves the converted text data and compares it to a database of previously collected fraudulent activity. This process utilizes AI algorithms to analyze and identify potential fraud in real time. In particular, it is designed to effectively detect specific phrases and contexts that may be associated with fraud.

[0400] If a fraudulent activity is deemed highly likely, the server controls the alarm system to send an alert to the user's family and pre-registered public institutions. This alert includes details about the potential fraud and the user's current location, allowing for a swift response.

[0401] As a concrete example of implementation, consider a scenario where a user receives a suspicious phone call and it is identified as a fraudulent billing attempt. In this case, the device monitors the conversation, processes the necessary information, and immediately warns the family, allowing the user to take steps to prevent becoming a victim of fraud.

[0402] In this way, this system can dynamically and quickly detect fraud using voice information and send alerts to relevant parties, thereby preventing fraud from occurring.

[0403] The following describes the processing flow.

[0404] Step 1:

[0405] The device monitors the sounds around the user in real time and collects them as audio data. Using an acoustic processing device, it reduces ambient noise while clearly capturing the content of conversations.

[0406] Step 2:

[0407] The terminal transmits the collected audio data to the data conversion unit. Here, the audio data is converted into text data using speech recognition software. The converted text is processed to accurately represent the conversation content without being affected by pronunciation quirks or noise.

[0408] Step 3:

[0409] The server receives the text data and applies an AI algorithm to compare it with a fraud database. The AI ​​uses machine learning models to identify unknown tactics as well as matching against known fraud patterns. At this stage, the likelihood of fraud is analyzed.

[0410] Step 4:

[0411] If the server determines that there is a high risk of fraud, it sends information to the alerting unit. The notification is then generated to include details of the suspected fraud and the user's location information.

[0412] Step 5:

[0413] The server sends alerts to contacts who need to be notified. The alerts are promptly sent via email or SMS to the user's family and designated public authorities.

[0414] Step 6:

[0415] The device, upon receiving an alarm, will provide the user with audio and visual warnings. This notification will inform the user of the potential for fraud and encourage them to take appropriate action.

[0416] These steps ensure the system functions in real time to protect users and prevent fraudulent activity.

[0417] (Example 1)

[0418] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0419] In modern society, individuals, including the elderly, are increasingly at risk of becoming victims of fraud. Fraud using voice, in particular, is easy to carry out and prone to causing harm. There is a need for real-time and reliable monitoring and notification methods to protect individuals from such fraud.

[0420] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0421] In this invention, the server includes means for collecting acoustic data, means for utilizing speech recognition technology to convert the acoustic data into text information, and means for analyzing the converted text information and identifying harmful acts using machine learning algorithms. This makes it possible to analyze user conversations in real time, quickly detect the risk of fraud, and appropriately send necessary notifications.

[0422] A "user" refers to a person whose voice data is monitored using the system.

[0423] A "terminal" refers to a device capable of collecting and analyzing acoustic data.

[0424] "Audio data" refers to information about audio waveforms collected by a device.

[0425] "Speech recognition technology" refers to a technical method for converting audio data into textual information.

[0426] "Textual information" refers to data in text format obtained using speech recognition technology.

[0427] A "machine learning algorithm" refers to a programmatic method for analyzing textual information and identifying harmful behavior.

[0428] An "alert" refers to a notification issued when a potential fraud is detected.

[0429] An "emergency contact" refers to a pre-registered individual or organization to which an alert is sent.

[0430] A "public service agency" refers to an organization formed to provide social safety and support.

[0431] This invention provides a system that enables early detection and notification of harmful acts using voice data. This system is based on three components: a terminal, a server, and a user.

[0432] The device is designed to collect audio from the user's surroundings. It incorporates a microphone system and constantly monitors the user's audio environment. The collected audio data is immediately converted into text using speech recognition software. To achieve high accuracy in this process, speech recognition technologies such as the Google Speech-to-Text API are used.

[0433] The server receives text information sent from the terminal and analyzes it using machine learning algorithms. Specifically, it utilizes generative AI models such as BERT and GPT models to detect potential harmful activities in the text data in real time. The server compares this information with a database of known harmful activities and promptly sends out a notification if fraud is suspected. The notification is sent through an alarm system, and the information is provided to emergency contacts and public service agencies.

[0434] This system allows users to take swift action when suspicious audio is detected. For example, if a user receives a suspicious phone call and it is determined to be a fraudulent billing attempt, the system will immediately alert family members and relevant parties, allowing the user to take appropriate action based on that alert.

[0435] An example of a prompt message would be, "Please describe the features of the fraud detection system that utilizes speech recognition technology." In this way, the present invention provides real-time detection and countermeasures for harmful acts by utilizing voice data.

[0436] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0437] Step 1:

[0438] The device collects audio data from the user's surroundings. Audio data input is obtained through a built-in microphone system. This audio data is recorded as a raw acoustic signal and then formatted into a specific format to facilitate processing in the next step. Furthermore, the recorded audio data is improved in quality through a noise reduction filter.

[0439] Step 2:

[0440] The device converts the collected audio data into text information. This stage utilizes speech recognition technologies such as the Google Speech-to-Text API. It receives audio data as input and generates text information (text data) as output. This conversion process involves analyzing the characteristics of the audio signal and selecting appropriate words. Finally, the audio conversation is output in a text format that is easy to understand.

[0441] Step 3:

[0442] The server processes text information received from the terminal. It takes text data as input and uses machine learning algorithms to detect potential harmful activities within it. Specifically, it uses generative AI models such as BERT and GPT to analyze specific phrases and contexts within the text. The output provides an assessment of the likelihood of fraud or harmful activity. This information serves as foundational data for the next step.

[0443] Step 4:

[0444] The server issues an alert if it determines that a scam is highly likely. Here, it uses the evaluation results obtained from step 3 as input to generate notifications for family members and public service agencies. The notifications include details of the fraudulent activity and the user's location information to facilitate a quick response to the emergency. As output, real-time alerts are sent.

[0445] Step 5:

[0446] The user receives an alert from the system and immediately recognizes the risk. Upon receiving the notification, the user can take appropriate action based on the information provided. At this stage, it is recommended that the user's family and related parties also cooperate to take action to protect the user from potential dangers.

[0447] (Application Example 1)

[0448] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0449] Currently, the number of cases in which users, including the elderly, fall victim to fraud is increasing, making fraud prevention a critical social issue. Traditional methods often result in delays in detecting fraud and insufficient response. Therefore, there is a need for a system that monitors the audio environment in real time and quickly detects and notifies users of potential fraud.

[0450] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0451] In this invention, the server includes acoustic processing means for monitoring the user's voice environment and collecting acoustic data, data conversion means for converting the collected acoustic data into text information, and analysis means for comparing the converted text information with existing fraud patterns to determine the likelihood of fraud. This makes it possible to monitor the likelihood of fraud in real time and issue warnings quickly.

[0452] "Acoustic processing means" refers to a device that continuously monitors the sound environment surrounding the user and collects necessary acoustic data.

[0453] A "data conversion means" is a device that converts collected acoustic data into textual information and generates text using high-precision speech recognition technology.

[0454] The "analysis tool" is a device that compares the converted text information with existing fraud patterns to determine the likelihood of fraud, and performs real-time analysis based on AI technology.

[0455] A "notification system" is a device that, when a potential scam is detected, issues a warning to pre-registered contacts, prompting them to take prompt action.

[0456] "Portable information devices" refer to portable information terminals such as smartphones and smart glasses, which serve as platforms that integrate sound processing capabilities.

[0457] To implement this invention, it is necessary to construct a system that mainly includes acoustic processing means, data conversion means, analysis means, and notification means. It is desirable that the system be integrated into a portable information device such as a smartphone or smart glasses.

[0458] The device uses acoustic processing to monitor audio data around the user in real time. This acoustic data is converted into highly accurate text information by a data conversion tool using the Google Cloud Speech-to-Text API. The converted text information is analyzed on a server using AI technologies such as TensorFlow, and the likelihood of fraud is determined by comparing it with existing fraud patterns.

[0459] If a user is deemed highly likely to be a scam, the server uses notification methods such as the Twilio API to send a warning to pre-registered contacts. This allows the user's family or caregivers to take immediate action.

[0460] For example, if a user receives a phone call from someone asking for their bank information, the device collects and analyzes the audio. If it determines that the call may be fraudulent, it immediately sends an alert to family members saying, "A suspicious call has been detected. Please check."

[0461] An example of a prompt message that would be effective is: "Please analyze the text of this audio and assess the likelihood of fraud. List any suspicious phrases or contexts and notify me of the results."

[0462] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0463] Step 1:

[0464] The terminal acquires audio data from the user's surroundings using acoustic processing equipment. The input is an audio signal, and the output is audio data. At this stage, the operation involves capturing the audio signal through the microphone.

[0465] Step 2:

[0466] The device converts the collected audio data into text using a data conversion method. The input is audio data, and the output is text data. This conversion uses the Google Cloud Speech-to-Text API to perform highly accurate speech recognition.

[0467] Step 3:

[0468] The server analyzes the text data received from the data transformation device using an analysis device. The input is text data, and the output is an evaluation result of whether or not it is potentially fraudulent. Specifically, it uses an AI model based on TensorFlow to compare the text data with existing fraud patterns.

[0469] Step 4:

[0470] If the server determines, based on the analysis results, that a scam is highly likely, it will send an alert to pre-registered contacts using a notification method. The input is the scam assessment result, and the output is a warning message. Here, the process of sending emails or SMS messages is performed using the Twilio API, etc.

[0471] Step 5:

[0472] The user's family and care staff receive warning messages from the device and take prompt action. The input is the warning message, and the output is the actual response action. Specific actions include contacting or visiting the user to ensure their safety.

[0473] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0474] This invention relates to a system that enables a user to quickly detect situations in which they are at risk of fraud and to take appropriate action. This system includes an emotion engine that has the capability to recognize the user's emotions.

[0475] The emotion engine is designed to analyze the user's emotions in real time from their voice. The device uses its acoustic processing unit to monitor the user's conversation and collect voice data, while the emotion engine extracts emotional data. The extracted emotional data is used to capture the user's stress level and signs of anxiety.

[0476] Once voice data is collected, the terminal converts it into text data using a data conversion unit. In parallel, the emotion engine records changes in emotional state and passes this data to a comparison device. The server uses the received text data and emotion data to compare and analyze them with existing fraud patterns and assess the likelihood of fraud.

[0477] The server comprehensively assesses the likelihood of fraud and the user's emotional state, and the alert system sets the alert level. Based on this assessment, the alert system sends notifications to the user's family and public institutions. The urgency of the alert is adjusted based on emotional data, so if there is a significant change in the user's emotions, immediate action is prompted.

[0478] For example, if a user comes into contact with a scammer over the phone, and their anxiety increases during the conversation, the emotion engine will immediately detect this change. This allows the system to add emotional information to its usual scam pattern detection and quickly issue alerts, enabling a faster response.

[0479] In this way, this system, equipped with an emotion engine, further enhances user safety by detecting potential fraud with high accuracy and issuing timely warnings.

[0480] The following describes the processing flow.

[0481] Step 1:

[0482] The device uses a microphone to monitor the user's conversation in real time and collect audio data. An acoustic processing unit reduces noise and ensures a clear audio signal.

[0483] Step 2:

[0484] The device inputs the collected audio data into an emotion engine, which analyzes parameters such as tone, pitch, and speed to identify the user's emotional state. In particular, it looks for signs of anxiety or stress.

[0485] Step 3:

[0486] The terminal simultaneously sends audio data to the data conversion unit, which converts the audio to text. Speech recognition technology is used to accurately transcribe the conversation into text.

[0487] Step 4:

[0488] The server receives emotional data from the emotion engine and text data from the data transformation unit. After receiving the data, it applies an AI algorithm to compare the emotional changes with known patterns of fraud.

[0489] Step 5:

[0490] The server assesses the risk of fraud and increases the risk level if the user is under high stress. It makes a comprehensive judgment on the likelihood of fraudulent activity.

[0491] Step 6:

[0492] The server, via its alarm system, sends alerts to the user's family and public institutions based on the detected fraud risk. The urgency of the alert is adjusted based on the degree of emotional change.

[0493] Step 7:

[0494] The device receives an alert and provides the user with a visual or auditory warning. This allows the user to recognize the situation and take appropriate action.

[0495] This process allows the system to take the user's emotional state into account and enhance its vigilance against fraud.

[0496] (Example 2)

[0497] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0498] Conventional fraud detection systems fail to take into account the user's emotional state, making it difficult to quickly and accurately assess the likelihood of fraud. Furthermore, there is a lack of information to enable users to take appropriate action in emergency situations when they are exposed to fraud risks.

[0499] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0500] In this invention, the server includes means for collecting acoustic signals and extracting emotional information, means for converting acoustic signals into textual information, and means for evaluating the possibility of fraud using the textual information and emotional information. This makes it possible to accurately grasp changes in the user's emotions, quickly assess the risk of fraud, and notify external organizations.

[0501] A "user" refers to an individual or organization that utilizes a system.

[0502] An "acoustic signal" is audio data generated by a user's conversation, and is an analog or digital sound waveform collected by a device such as a microphone.

[0503] "Device means" refers to physical or virtual devices or equipment used to achieve a specific function.

[0504] "Emotional information" refers to data extracted from acoustic signals, and includes indicators and categories that show the user's emotional state.

[0505] "Textual information" refers to text data obtained by converting acoustic signals, and is a string of characters that represents the content of a conversation.

[0506] A "fraud pattern" is a dataset that compiles the characteristics and methods of fraudulent activities reported in the past, and is used to evaluate the potential of new frauds.

[0507] "External organization" refers to an external group, individual, or public institution to which warnings or notifications are sent.

[0508] The embodiments for carrying out the present invention will be described below.

[0509] The device includes an acoustic processing unit for monitoring user conversations in real time. This unit uses a microphone to collect acoustic signals and converts them into digital signals. This digitized audio data is sent to an emotion engine to extract the user's emotions. The emotion engine uses machine learning algorithms to determine emotional information from the acoustic signals in real time. It is also possible to utilize external audio processing platforms (such as Amazon Web Services or Google Cloud Platform's speech analysis APIs) for this process.

[0510] The collected audio data is converted into text information via speech recognition software. For example, the Microsoft Azure speech recognition API can be used for this conversion.

[0511] The server uses textual and sentimental information received from the terminal to compare and analyze it against existing fraud patterns. This process utilizes generative AI models and natural language processing techniques to assess the likelihood of fraud. Based on the evaluated data, the server sets a warning level and sends necessary notifications to external organizations. In this process, it is conceivable that warnings would be sent via SMS or email using communication APIs such as Twilio.

[0512] For example, if a user receives a potentially fraudulent phone call, the device detects changes in the user's voice during the conversation, and the emotion engine immediately detects an increase in anxiety. The server analyzes the emotional changes and text information, and if it assesses a high probability of fraud, it sets an "urgent" alert level and sends a notification to the user's family and relevant organizations.

[0513] Examples of input prompts for a generative AI model:

[0514] "How does the emotion engine respond when a user receives a potentially fraudulent phone call? Please describe in detail the system's response based on emotion detection."

[0515] In this way, the system enables highly accurate fraud detection based on changes in the user's emotions and enhances user safety through timely alarm notifications.

[0516] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0517] Step 1:

[0518] The terminal collects acoustic signals in real time using a microphone to receive user conversations. The input for this step is ambient noise and user speech. The terminal converts the acoustic signals into digital data and performs signal processing such as noise filtering to clearly extract the user's voice. The output is clear digital audio data.

[0519] Step 2:

[0520] The device sends the collected digital audio data to the emotion engine, which then extracts emotional information. The input for this step is framed audio data. The device analyzes features such as tone, volume changes, and speed of the audio and uses a machine learning algorithm to classify the user's emotional state. The output is data indicating the user's emotion (e.g., "relieved," "tense," "anxious").

[0521] Step 3:

[0522] The device uses speech recognition software to convert digital audio data into text. The input for this step is pre-processed digital audio data. The device converts the audio to text, for example, using the Google Speech-to-Text API. The resulting text data represents the conversation content as written text.

[0523] Step 4:

[0524] The server analyzes the likelihood of fraud using text and sentiment information received from the terminal. The input for this step is text data and sentiment data. The server uses natural language processing techniques to analyze the text information and compare it to existing fraud patterns in the database. It also uses a generative AI model to predict unknown fraud patterns. The output is evaluation data indicating the likelihood and confidence level of fraud.

[0525] Step 5:

[0526] The server sets the alert level and sends out notifications. The input for this step is fraud rating data and user sentiment data. Based on the rating data, the server determines an alert level such as "Low," "Medium," "High," or "Emergency." Based on this, it transmits alerts to family members or public institutions using communication APIs such as Twilio. The output is the set alert level and a message notifying them of this.

[0527] Each of these processing steps allows the system to capture changes in the user's emotions while efficiently detecting and notifying them of potential fraud.

[0528] (Application Example 2)

[0529] Next, we will explain Application Example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0530] Conventional fraud detection systems rely solely on comparing voice data and fail to consider changes in the user's emotions, making it difficult to accurately determine the likelihood of fraud. Furthermore, they may not adequately reflect the user's anxiety or stress levels, potentially leading to delays in issuing timely warnings.

[0531] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0532] In this invention, the server includes data conversion means for converting voice information into text, data comparison means for comparing the converted text with fraud patterns, emotion analysis means for analyzing emotions from voice in real time, and alarm issuing means for issuing notifications based on the likelihood of fraud and emotion analysis. This enables highly accurate fraud detection and timely alarm issuance that takes into account the user's emotional state.

[0533] A "user" is an individual or legal entity that uses the system to make calls or communicate.

[0534] A "telephone call" is a real-time communication method that uses voice to exchange information.

[0535] "Audio information" refers to audio data that includes the words and utterances spoken by the user.

[0536] "Acoustic analysis means" refers to a device or system for acquiring audio information and performing necessary processing.

[0537] "Data conversion means" refers to a device or program that has the function of converting audio information into text data.

[0538] A "data comparison means" is a device or program that compares converted text with existing patterns and evaluates the degree of matching or similarity.

[0539] An "emotion analysis device" is a device or system that analyzes a user's emotions in real time from their voice and evaluates the degree of stress and anxiety.

[0540] A "fraud pattern" is data that shows typical combinations of behaviors and expressions based on past fraudulent activities.

[0541] An "alarm notification device" is a device or system that has the function of notifying in a pre-set manner when it detects the possibility of fraud.

[0542] A "portable communication device" is a portable device that has communication capabilities for sending and receiving voice and data while on the move.

[0543] A "server" is a computer system or device that processes data over a network and provides services to clients.

[0544] The system for realizing this invention includes acoustic analysis means, data conversion means, data comparison means, emotion analysis means, and alarm issuing means. This system enables faster and more accurate detection of potential fraud by reflecting the user's emotional state in real time in conventional fraud detection technology.

[0545] The acoustic analysis system uses the microphone of a mobile communication device to collect user voice information. This collection process uses speech analysis software, such as the Google Cloud Speech-to-Text API, to convert the voice data into text data. The converted text data is then sent to a server.

[0546] The server first uses a data comparison tool to match text data against existing fraud patterns. Next, an emotion analysis tool analyzes the user's emotions extracted from the audio information and evaluates their stress and anxiety levels. This analysis utilizes emotion analysis models based on TensorFlow or PyTorch.

[0547] After integrating the user's emotional state with the results of data comparison, the server uses an alarm system to issue a warning if fraud is suspected. This warning is sent to the user themselves, or, in some cases, to appropriate contacts such as family members or public institutions. The urgency of the notification is flexibly adjusted according to the magnitude of the user's emotional change, enabling a quick response.

[0548] For example, if an unstable or high-stress state is detected when a user provides credit card information in an e-commerce transaction, an alert will be issued immediately. This operation strengthens defenses against fraudulent activity and improves user safety.

[0549] An example of a prompt might be: "Design a voice sentiment analysis app that detects potential fraud during a call. Generate a prototype that issues a warning when the user becomes anxious, and explain how the system works."

[0550] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0551] Step 1:

[0552] The terminal collects user voice information in real time using the microphone of the mobile communication device. The input is raw voice data, which is transmitted directly to the acoustic analysis function. Using the acoustic analysis means, noise in the voice information is removed and output as clear voice data.

[0553] Step 2:

[0554] The terminal inputs clear audio data obtained by the acoustic analysis means into the data conversion means. Here, the Google Cloud Speech-to-Text API is used to convert the audio data into text data. The output is the converted text data.

[0555] Step 3:

[0556] The server inputs the text data sent from the terminal into a data comparison device. Here, the text data is compared to predefined fraud patterns, and a match or similarity is evaluated. A fraud probability score is generated as output.

[0557] Step 4:

[0558] The terminal then again supplies the user's voice information in real time, this time to the emotion analysis system. Emotional data is extracted from the voice using an emotion analysis model based on TensorFlow or PyTorch. The input is raw voice data, and the output is the user's emotional state (e.g., stress level, degree of anxiety).

[0559] Step 5:

[0560] The server integrates the fraud probability score from Step 3 and the emotional state from Step 4 to make an overall judgment. Here, it performs data calculations to determine the urgency of a fraud alert. The output is notification data that includes the urgency for issuing an alert.

[0561] Step 6:

[0562] The server issues notifications based on the generated urgency level through an alarm system. The input is notification data, and the output is an alarm sent to the user or a designated contact (e.g., family, public facilities). The content of the alarm is customized according to the likelihood of fraud and changes in the user's emotional state.

[0563] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0564] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0565] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0566] [Fourth Embodiment]

[0567] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0568] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0569] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0570] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0571] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0572] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0573] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0574] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0575] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0576] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0577] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0578] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0579] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0580] This invention relates to an audio data monitoring system for protecting users, including the elderly, from fraud. The system consists of an acoustic processing unit, a data conversion unit, a comparison unit, and an alarm issuing unit.

[0581] First, the device constantly monitors the sounds around the user and uses an acoustic processing unit to collect conversations. Once the audio data is collected, a data conversion unit within the system receives it and converts it from audio to text data. During this conversion process, speech recognition technology generates highly accurate text.

[0582] Next, the server retrieves the converted text data and compares it to a database of previously collected fraudulent activity. This process utilizes AI algorithms to analyze and identify potential fraud in real time. In particular, it is designed to effectively detect specific phrases and contexts that may be associated with fraud.

[0583] If a fraudulent activity is deemed highly likely, the server controls the alarm system to send an alert to the user's family and pre-registered public institutions. This alert includes details about the potential fraud and the user's current location, allowing for a swift response.

[0584] As a concrete example of implementation, consider a scenario where a user receives a suspicious phone call and it is identified as a fraudulent billing attempt. In this case, the device monitors the conversation, processes the necessary information, and immediately warns the family, allowing the user to take steps to prevent becoming a victim of fraud.

[0585] In this way, this system can dynamically and quickly detect fraud using voice information and send alerts to relevant parties, thereby preventing fraud from occurring.

[0586] The following describes the processing flow.

[0587] Step 1:

[0588] The device monitors the sounds around the user in real time and collects them as audio data. Using an acoustic processing device, it reduces ambient noise while clearly capturing the content of conversations.

[0589] Step 2:

[0590] The terminal transmits the collected audio data to the data conversion unit. Here, the audio data is converted into text data using speech recognition software. The converted text is processed to accurately represent the conversation content without being affected by pronunciation quirks or noise.

[0591] Step 3:

[0592] The server receives the text data and applies an AI algorithm to compare it with a fraud database. The AI ​​uses machine learning models to identify unknown tactics as well as matching against known fraud patterns. At this stage, the likelihood of fraud is analyzed.

[0593] Step 4:

[0594] If the server determines that there is a high risk of fraud, it sends information to the alerting unit. The notification is then generated to include details of the suspected fraud and the user's location information.

[0595] Step 5:

[0596] The server sends alerts to contacts who need to be notified. The alerts are promptly sent via email or SMS to the user's family and designated public authorities.

[0597] Step 6:

[0598] The device, upon receiving an alarm, will provide the user with audio and visual warnings. This notification will inform the user of the potential for fraud and encourage them to take appropriate action.

[0599] These steps ensure the system functions in real time to protect users and prevent fraudulent activity.

[0600] (Example 1)

[0601] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0602] In modern society, individuals, including the elderly, are increasingly at risk of becoming victims of fraud. Fraud using voice, in particular, is easy to carry out and prone to causing harm. There is a need for real-time and reliable monitoring and notification methods to protect individuals from such fraud.

[0603] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0604] In this invention, the server includes means for collecting acoustic data, means for utilizing speech recognition technology to convert the acoustic data into text information, and means for analyzing the converted text information and identifying harmful acts using machine learning algorithms. This makes it possible to analyze user conversations in real time, quickly detect the risk of fraud, and appropriately send necessary notifications.

[0605] A "user" refers to a person whose voice data is monitored using the system.

[0606] A "terminal" refers to a device capable of collecting and analyzing acoustic data.

[0607] "Audio data" refers to information about audio waveforms collected by a device.

[0608] "Speech recognition technology" refers to a technical method for converting audio data into textual information.

[0609] "Textual information" refers to data in text format obtained using speech recognition technology.

[0610] A "machine learning algorithm" refers to a programmatic method for analyzing textual information and identifying harmful behavior.

[0611] An "alert" refers to a notification issued when a potential fraud is detected.

[0612] An "emergency contact" refers to a pre-registered individual or organization to which an alert is sent.

[0613] A "public service agency" refers to an organization formed to provide social safety and support.

[0614] This invention provides a system that enables early detection and notification of harmful acts using voice data. This system is based on three components: a terminal, a server, and a user.

[0615] The device is designed to collect audio from the user's surroundings. It incorporates a microphone system and constantly monitors the user's audio environment. The collected audio data is immediately converted into text using speech recognition software. To achieve high accuracy in this process, speech recognition technologies such as the Google Speech-to-Text API are used.

[0616] The server receives text information sent from the terminal and analyzes it using machine learning algorithms. Specifically, it utilizes generative AI models such as BERT and GPT models to detect potential harmful activities in the text data in real time. The server compares this information with a database of known harmful activities and promptly sends out a notification if fraud is suspected. The notification is sent through an alarm system, and the information is provided to emergency contacts and public service agencies.

[0617] This system allows users to take swift action when suspicious audio is detected. For example, if a user receives a suspicious phone call and it is determined to be a fraudulent billing attempt, the system will immediately alert family members and relevant parties, allowing the user to take appropriate action based on that alert.

[0618] An example of a prompt message would be, "Please describe the features of the fraud detection system that utilizes speech recognition technology." In this way, the present invention provides real-time detection and countermeasures for harmful acts by utilizing voice data.

[0619] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0620] Step 1:

[0621] The device collects audio data from the user's surroundings. Audio data input is obtained through a built-in microphone system. This audio data is recorded as a raw acoustic signal and then formatted into a specific format to facilitate processing in the next step. Furthermore, the recorded audio data is improved in quality through a noise reduction filter.

[0622] Step 2:

[0623] The device converts the collected audio data into text information. This stage utilizes speech recognition technologies such as the Google Speech-to-Text API. It receives audio data as input and generates text information (text data) as output. This conversion process involves analyzing the characteristics of the audio signal and selecting appropriate words. Finally, the audio conversation is output in a text format that is easy to understand.

[0624] Step 3:

[0625] The server processes text information received from the terminal. It takes text data as input and uses machine learning algorithms to detect potential harmful activities within it. Specifically, it uses generative AI models such as BERT and GPT to analyze specific phrases and contexts within the text. The output provides an assessment of the likelihood of fraud or harmful activity. This information serves as foundational data for the next step.

[0626] Step 4:

[0627] The server issues an alert if it determines that a scam is highly likely. Here, it uses the evaluation results obtained from step 3 as input to generate notifications for family members and public service agencies. The notifications include details of the fraudulent activity and the user's location information to facilitate a quick response to the emergency. As output, real-time alerts are sent.

[0628] Step 5:

[0629] The user receives an alert from the system and immediately recognizes the risk. Upon receiving the notification, the user can take appropriate action based on the information provided. At this stage, it is recommended that the user's family and related parties also cooperate to take action to protect the user from potential dangers.

[0630] (Application Example 1)

[0631] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0632] Currently, the number of cases in which users, including the elderly, fall victim to fraud is increasing, making fraud prevention a critical social issue. Traditional methods often result in delays in detecting fraud and insufficient response. Therefore, there is a need for a system that monitors the audio environment in real time and quickly detects and notifies users of potential fraud.

[0633] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0634] In this invention, the server includes acoustic processing means for monitoring the user's voice environment and collecting acoustic data, data conversion means for converting the collected acoustic data into text information, and analysis means for comparing the converted text information with existing fraud patterns to determine the likelihood of fraud. This makes it possible to monitor the likelihood of fraud in real time and issue warnings quickly.

[0635] "Acoustic processing means" refers to a device that continuously monitors the sound environment surrounding the user and collects necessary acoustic data.

[0636] A "data conversion means" is a device that converts collected acoustic data into textual information and generates text using high-precision speech recognition technology.

[0637] The "analysis tool" is a device that compares the converted text information with existing fraud patterns to determine the likelihood of fraud, and performs real-time analysis based on AI technology.

[0638] A "notification system" is a device that, when a potential scam is detected, issues a warning to pre-registered contacts, prompting them to take prompt action.

[0639] "Portable information devices" refer to portable information terminals such as smartphones and smart glasses, which serve as platforms that integrate sound processing capabilities.

[0640] To implement this invention, it is necessary to construct a system that mainly includes acoustic processing means, data conversion means, analysis means, and notification means. It is desirable that the system be integrated into a portable information device such as a smartphone or smart glasses.

[0641] The device uses acoustic processing to monitor audio data around the user in real time. This acoustic data is converted into highly accurate text information by a data conversion tool using the Google Cloud Speech-to-Text API. The converted text information is analyzed on a server using AI technologies such as TensorFlow, and the likelihood of fraud is determined by comparing it with existing fraud patterns.

[0642] If a user is deemed highly likely to be a scam, the server uses notification methods such as the Twilio API to send a warning to pre-registered contacts. This allows the user's family or caregivers to take immediate action.

[0643] For example, if a user receives a phone call from someone asking for their bank information, the device collects and analyzes the audio. If it determines that the call may be fraudulent, it immediately sends an alert to family members saying, "A suspicious call has been detected. Please check."

[0644] An example of a prompt message that would be effective is: "Please analyze the text of this audio and assess the likelihood of fraud. List any suspicious phrases or contexts and notify me of the results."

[0645] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0646] Step 1:

[0647] The terminal acquires audio data from the user's surroundings using acoustic processing equipment. The input is an audio signal, and the output is audio data. At this stage, the operation involves capturing the audio signal through the microphone.

[0648] Step 2:

[0649] The device converts the collected audio data into text using a data conversion method. The input is audio data, and the output is text data. This conversion uses the Google Cloud Speech-to-Text API to perform highly accurate speech recognition.

[0650] Step 3:

[0651] The server analyzes the text data received from the data transformation device using an analysis device. The input is text data, and the output is an evaluation result of whether or not it is potentially fraudulent. Specifically, it uses an AI model based on TensorFlow to compare the text data with existing fraud patterns.

[0652] Step 4:

[0653] If the server determines, based on the analysis results, that a scam is highly likely, it will send an alert to pre-registered contacts using a notification method. The input is the scam assessment result, and the output is a warning message. Here, the process of sending emails or SMS messages is performed using the Twilio API, etc.

[0654] Step 5:

[0655] The user's family and care staff receive warning messages from the device and take prompt action. The input is the warning message, and the output is the actual response action. Specific actions include contacting or visiting the user to ensure their safety.

[0656] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0657] This invention relates to a system that enables a user to quickly detect situations in which they are at risk of fraud and to take appropriate action. This system includes an emotion engine that has the capability to recognize the user's emotions.

[0658] The emotion engine is designed to analyze the user's emotions in real time from their voice. The device uses its acoustic processing unit to monitor the user's conversation and collect voice data, while the emotion engine extracts emotional data. The extracted emotional data is used to capture the user's stress level and signs of anxiety.

[0659] Once voice data is collected, the terminal converts it into text data using a data conversion unit. In parallel, the emotion engine records changes in emotional state and passes this data to a comparison device. The server uses the received text data and emotion data to compare and analyze them with existing fraud patterns and assess the likelihood of fraud.

[0660] The server comprehensively assesses the likelihood of fraud and the user's emotional state, and the alert system sets the alert level. Based on this assessment, the alert system sends notifications to the user's family and public institutions. The urgency of the alert is adjusted based on emotional data, so if there is a significant change in the user's emotions, immediate action is prompted.

[0661] For example, if a user comes into contact with a scammer over the phone, and their anxiety increases during the conversation, the emotion engine will immediately detect this change. This allows the system to add emotional information to its usual scam pattern detection and quickly issue alerts, enabling a faster response.

[0662] In this way, this system, equipped with an emotion engine, further enhances user safety by detecting potential fraud with high accuracy and issuing timely warnings.

[0663] The following describes the processing flow.

[0664] Step 1:

[0665] The device uses a microphone to monitor the user's conversation in real time and collect audio data. An acoustic processing unit reduces noise and ensures a clear audio signal.

[0666] Step 2:

[0667] The device inputs the collected audio data into an emotion engine, which analyzes parameters such as tone, pitch, and speed to identify the user's emotional state. In particular, it looks for signs of anxiety or stress.

[0668] Step 3:

[0669] The terminal simultaneously sends audio data to the data conversion unit, which converts the audio to text. Speech recognition technology is used to accurately transcribe the conversation into text.

[0670] Step 4:

[0671] The server receives emotional data from the emotion engine and text data from the data transformation unit. After receiving the data, it applies an AI algorithm to compare the emotional changes with known patterns of fraud.

[0672] Step 5:

[0673] The server assesses the risk of fraud and increases the risk level if the user is under high stress. It makes a comprehensive judgment on the likelihood of fraudulent activity.

[0674] Step 6:

[0675] The server, via its alarm system, sends alerts to the user's family and public institutions based on the detected fraud risk. The urgency of the alert is adjusted based on the degree of emotional change.

[0676] Step 7:

[0677] The device receives an alert and provides the user with a visual or auditory warning. This allows the user to recognize the situation and take appropriate action.

[0678] This process allows the system to take the user's emotional state into account and enhance its vigilance against fraud.

[0679] (Example 2)

[0680] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0681] Conventional fraud detection systems fail to take into account the user's emotional state, making it difficult to quickly and accurately assess the likelihood of fraud. Furthermore, there is a lack of information to enable users to take appropriate action in emergency situations when they are exposed to fraud risks.

[0682] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0683] In this invention, the server includes means for collecting acoustic signals and extracting emotional information, means for converting acoustic signals into textual information, and means for evaluating the possibility of fraud using the textual information and emotional information. This makes it possible to accurately grasp changes in the user's emotions, quickly assess the risk of fraud, and notify external organizations.

[0684] A "user" refers to an individual or organization that utilizes a system.

[0685] An "acoustic signal" is audio data generated by a user's conversation, and is an analog or digital sound waveform collected by a device such as a microphone.

[0686] "Device means" refers to physical or virtual devices or equipment used to achieve a specific function.

[0687] "Emotional information" refers to data extracted from acoustic signals, and includes indicators and categories that show the user's emotional state.

[0688] "Textual information" refers to text data obtained by converting acoustic signals, and is a string of characters that represents the content of a conversation.

[0689] A "fraud pattern" is a dataset that compiles the characteristics and methods of fraudulent activities reported in the past, and is used to evaluate the potential of new frauds.

[0690] "External organization" refers to an external group, individual, or public institution to which warnings or notifications are sent.

[0691] The embodiments for carrying out the present invention will be described below.

[0692] The device includes an acoustic processing unit for monitoring user conversations in real time. This unit uses a microphone to collect acoustic signals and converts them into digital signals. This digitized audio data is sent to an emotion engine to extract the user's emotions. The emotion engine uses machine learning algorithms to determine emotional information from the acoustic signals in real time. It is also possible to utilize external audio processing platforms (such as Amazon Web Services or Google Cloud Platform's speech analysis APIs) for this process.

[0693] The collected audio data is converted into text information via speech recognition software. For example, the Microsoft Azure speech recognition API can be used for this conversion.

[0694] The server uses textual and sentimental information received from the terminal to compare and analyze it against existing fraud patterns. This process utilizes generative AI models and natural language processing techniques to assess the likelihood of fraud. Based on the evaluated data, the server sets a warning level and sends necessary notifications to external organizations. In this process, it is conceivable that warnings would be sent via SMS or email using communication APIs such as Twilio.

[0695] For example, if a user receives a potentially fraudulent phone call, the device detects changes in the user's voice during the conversation, and the emotion engine immediately detects an increase in anxiety. The server analyzes the emotional changes and text information, and if it assesses a high probability of fraud, it sets an "urgent" alert level and sends a notification to the user's family and relevant organizations.

[0696] Examples of input prompts for a generative AI model:

[0697] "How does the emotion engine respond when a user receives a potentially fraudulent phone call? Please describe in detail the system's response based on emotion detection."

[0698] In this way, the system enables highly accurate fraud detection based on changes in the user's emotions and enhances user safety through timely alarm notifications.

[0699] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0700] Step 1:

[0701] The terminal collects acoustic signals in real time using a microphone to receive user conversations. The input for this step is ambient noise and user speech. The terminal converts the acoustic signals into digital data and performs signal processing such as noise filtering to clearly extract the user's voice. The output is clear digital audio data.

[0702] Step 2:

[0703] The device sends the collected digital audio data to the emotion engine, which then extracts emotional information. The input for this step is framed audio data. The device analyzes features such as tone, volume changes, and speed of the audio and uses a machine learning algorithm to classify the user's emotional state. The output is data indicating the user's emotion (e.g., "relieved," "tense," "anxious").

[0704] Step 3:

[0705] The device uses speech recognition software to convert digital audio data into text. The input for this step is pre-processed digital audio data. The device converts the audio to text, for example, using the Google Speech-to-Text API. The resulting text data represents the conversation content as written text.

[0706] Step 4:

[0707] The server analyzes the likelihood of fraud using text and sentiment information received from the terminal. The input for this step is text data and sentiment data. The server uses natural language processing techniques to analyze the text information and compare it to existing fraud patterns in the database. It also uses a generative AI model to predict unknown fraud patterns. The output is evaluation data indicating the likelihood and confidence level of fraud.

[0708] Step 5:

[0709] The server sets the alert level and sends out notifications. The input for this step is fraud rating data and user sentiment data. Based on the rating data, the server determines an alert level such as "Low," "Medium," "High," or "Emergency." Based on this, it transmits alerts to family members or public institutions using communication APIs such as Twilio. The output is the set alert level and a message notifying them of this.

[0710] Each of these processing steps allows the system to capture changes in the user's emotions while efficiently detecting and notifying them of potential fraud.

[0711] (Application Example 2)

[0712] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0713] Conventional fraud detection systems rely solely on comparing voice data and fail to consider changes in the user's emotions, making it difficult to accurately determine the likelihood of fraud. Furthermore, they may not adequately reflect the user's anxiety or stress levels, potentially leading to delays in issuing timely warnings.

[0714] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0715] In this invention, the server includes data conversion means for converting voice information into text, data comparison means for comparing the converted text with fraud patterns, emotion analysis means for analyzing emotions from voice in real time, and alarm issuing means for issuing notifications based on the likelihood of fraud and emotion analysis. This enables highly accurate fraud detection and timely alarm issuance that takes into account the user's emotional state.

[0716] A "user" is an individual or legal entity that uses the system to make calls or communicate.

[0717] A "telephone call" is a real-time communication method that uses voice to exchange information.

[0718] "Audio information" refers to audio data that includes the words and utterances spoken by the user.

[0719] "Acoustic analysis means" refers to a device or system for acquiring audio information and performing necessary processing.

[0720] "Data conversion means" refers to a device or program that has the function of converting audio information into text data.

[0721] A "data comparison means" is a device or program that compares converted text with existing patterns and evaluates the degree of matching or similarity.

[0722] An "emotion analysis device" is a device or system that analyzes a user's emotions in real time from their voice and evaluates the degree of stress and anxiety.

[0723] A "fraud pattern" is data that shows typical combinations of behaviors and expressions based on past fraudulent activities.

[0724] An "alarm notification device" is a device or system that has the function of notifying in a pre-set manner when it detects the possibility of fraud.

[0725] A "portable communication device" is a portable device that has communication capabilities for sending and receiving voice and data while on the move.

[0726] A "server" is a computer system or device that processes data over a network and provides services to clients.

[0727] The system for realizing this invention includes acoustic analysis means, data conversion means, data comparison means, emotion analysis means, and alarm issuing means. This system enables faster and more accurate detection of potential fraud by reflecting the user's emotional state in real time in conventional fraud detection technology.

[0728] The acoustic analysis system uses the microphone of a mobile communication device to collect user voice information. This collection process uses speech analysis software, such as the Google Cloud Speech-to-Text API, to convert the voice data into text data. The converted text data is then sent to a server.

[0729] The server first uses a data comparison tool to match text data against existing fraud patterns. Next, an emotion analysis tool analyzes the user's emotions extracted from the audio information and evaluates their stress and anxiety levels. This analysis utilizes emotion analysis models based on TensorFlow or PyTorch.

[0730] After integrating the user's emotional state with the results of data comparison, the server uses an alarm system to issue a warning if fraud is suspected. This warning is sent to the user themselves, or, in some cases, to appropriate contacts such as family members or public institutions. The urgency of the notification is flexibly adjusted according to the magnitude of the user's emotional change, enabling a quick response.

[0731] For example, if an unstable or high-stress state is detected when a user provides credit card information in an e-commerce transaction, an alert will be issued immediately. This operation strengthens defenses against fraudulent activity and improves user safety.

[0732] An example of a prompt might be: "Design a voice sentiment analysis app that detects potential fraud during a call. Generate a prototype that issues a warning when the user becomes anxious, and explain how the system works."

[0733] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0734] Step 1:

[0735] The terminal collects user voice information in real time using the microphone of the mobile communication device. The input is raw voice data, which is transmitted directly to the acoustic analysis function. Using the acoustic analysis means, noise in the voice information is removed and output as clear voice data.

[0736] Step 2:

[0737] The terminal inputs clear audio data obtained by the acoustic analysis means into the data conversion means. Here, the Google Cloud Speech-to-Text API is used to convert the audio data into text data. The output is the converted text data.

[0738] Step 3:

[0739] The server inputs the text data sent from the terminal into a data comparison device. Here, the text data is compared to predefined fraud patterns, and a match or similarity is evaluated. A fraud probability score is generated as output.

[0740] Step 4:

[0741] The terminal then again supplies the user's voice information in real time, this time to the emotion analysis system. Emotional data is extracted from the voice using an emotion analysis model based on TensorFlow or PyTorch. The input is raw voice data, and the output is the user's emotional state (e.g., stress level, degree of anxiety).

[0742] Step 5:

[0743] The server integrates the fraud probability score from Step 3 and the emotional state from Step 4 to make an overall judgment. Here, it performs data calculations to determine the urgency of a fraud alert. The output is notification data that includes the urgency for issuing an alert.

[0744] Step 6:

[0745] The server issues notifications based on the generated urgency level through an alarm system. The input is notification data, and the output is an alarm sent to the user or a designated contact (e.g., family, public facilities). The content of the alarm is customized according to the likelihood of fraud and changes in the user's emotional state.

[0746] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0747] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0748] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0749] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0750] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0751] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0752] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0753] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0754] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0755] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0756] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0757] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0758] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0759] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[0760] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0761] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0762] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0763] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0764] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0765] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0766] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0767] The following is further disclosed regarding the embodiments described above.

[0768] (Claim 1)

[0769] An acoustic processing device that monitors user conversations and collects audio data,

[0770] A data conversion unit that converts audio data into text,

[0771] A comparison device that compares the converted text with existing fraud patterns,

[0772] An alarm system that detects potential fraud and issues notifications,

[0773] A system that includes this.

[0774] (Claim 2)

[0775] The system according to claim 1, wherein the alarm transmitting unit transmits an alert to family members or public institutions.

[0776] (Claim 3)

[0777] The system according to claim 1, wherein the sound processing device is incorporated into a portable information terminal.

[0778] "Example 1"

[0779] (Claim 1)

[0780] A terminal is provided to collect and analyze the user's voice, and means are provided to collect acoustic data.

[0781] A method of using speech recognition technology to convert audio data into text information,

[0782] A means of analyzing converted text information and identifying harmful acts using machine learning algorithms,

[0783] A means of quickly issuing an alert and providing information to registered contacts,

[0784] A system that includes this.

[0785] (Claim 2)

[0786] The system according to claim 1, which transmits an alert to a pre-registered emergency contact or public service agency.

[0787] (Claim 3)

[0788] The system according to claim 1, wherein the acoustic data acquisition means is integrated into a portable electronic device.

[0789] "Application Example 1"

[0790] (Claim 1)

[0791] A sound processing means that monitors the user's voice environment and collects sound data,

[0792] A data conversion means for converting collected acoustic data into textual information,

[0793] An analytical method for comparing converted text information with existing fraud patterns to determine the likelihood of fraud,

[0794] A notification system that sends a warning to pre-registered contacts when a potential scam is detected,

[0795] A system that includes this.

[0796] (Claim 2)

[0797] The system according to claim 1, wherein the notification means transmits a warning to a close relative or public institution.

[0798] (Claim 3)

[0799] The system according to claim 1, wherein the sound processing means is integrated into a portable information device.

[0800] "Example 2 of combining an emotion engine"

[0801] (Claim 1)

[0802] A device for monitoring user conversations and collecting acoustic signals,

[0803] A means of extracting emotional information from collected acoustic signals,

[0804] A means of converting acoustic signals into textual information,

[0805] A means of evaluating the likelihood of fraud by comparing converted textual and emotional information with existing fraud patterns,

[0806] A means of detecting potential fraud and sending notifications to external organizations,

[0807] A system that includes this.

[0808] (Claim 2)

[0809] The system according to claim 1, wherein the alarm issuing means transmits a warning to an individual or organization.

[0810] (Claim 3)

[0811] The system according to claim 1, wherein the acoustic signal acquisition device is incorporated into a portable information device.

[0812] "Application example 2 when combining with an emotional engine"

[0813] (Claim 1)

[0814] An acoustic analysis means for monitoring user calls and collecting audio information,

[0815] A data conversion means for converting audio information into text,

[0816] A data comparison method that compares the converted text with a predefined fraud pattern,

[0817] A sentiment analysis method that analyzes emotions from the user's voice in real time,

[0818] An alarm system that issues notifications based on the likelihood of fraud and sentiment analysis,

[0819] A system that includes this.

[0820] (Claim 2)

[0821] The system according to claim 1, wherein the alarm transmission means transmits an alert to a family member or a public facility.

[0822] (Claim 3)

[0823] The system according to claim 1, wherein the acoustic analysis means is incorporated into a portable communication device. [Explanation of symbols]

[0824] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A sound processing means that monitors the user's voice environment and collects sound data, A data conversion means for converting collected acoustic data into textual information, An analytical method for comparing converted text information with existing fraud patterns to determine the likelihood of fraud, A notification system that sends a warning to pre-registered contacts when a potential scam is detected, A system that includes this.

2. The system according to claim 1, wherein the notification means transmits a warning to a close relative or a public institution.

3. The system according to claim 1, wherein the sound processing means is integrated into a portable information device.

Images